Slashdot Mirror
.75 GHz Athlon Released
News.com is reporting that AMD has released a new 750 MHz Athlon. The chip is quite pricey ($800 in lots of 1000), but should be available before the year is over. Jerry Sanders says AMD is having a strong quarter. Cnnfn.com also has an article about the release of the chip, and also mentions that a 533 MHz K6-2 was released.
A 750 Mhz Athlon cannot be much better than a 733 Mhz Athlon...? Surely the huge price premium is more to make a profit for those k-rad hax0rs who can go and measure their processor speed in gigahertz... AFAIK there are no > 700 Mhz CPUs available in the UK, so I can't get one for the moment, but when they do come out, will there be a practical use for them?
Since I didn't see any mention that these parts were coming from Dresden, does anyone know if that means Austin has already switched over to the 0.18um process? I was under the impression that they were going to ramp up Dresden and then take Austin down for the switch.
I'm all for technology advancing, really. Everything's getting smaller faster "better", and that's just beautiful. But this really screws up bragging. It just sounds weird to say "Yeah, I just bought a P5 1.25" "Pfft, I've got a 1.3, loser"
You better believe it.
At least strong enough to produce a good TV commercial where because some guy doesn't know the Athalon is faster than the Pentium III, he causes a train to smash into the building he's working in.
Although they can't be doing that well, since I've only seen it once as opposed to the millions and millions of times Intel commercials have aired.
Glückwünsche, haben Sie Slashdot ermordet, indem Sie zum korporativen Druck beugten und Subskriptionen einlei
The article makes a very minor mistake. They imply that the speed advantage has something to do with clock-rate, and of course, it has nothing to do with that. Other than that, it's a good article, though I'm not sure that the Intel stock price had anything to do with this development DIRECTLY.
Hey everyone,
.18u process. Yes, that means less heat and better OC possibilities abound!!
Just to let everyone know that the exciting thing about this chip is that it is made using the new
I may be just me, but I am waiting to see what KryoTech can pull off with this baby!
TheJet
The "Top 10" Reasons to procrastinate:
10.
Intel has had a close-to-be monopoly for all too long, and AMD is the best chanse we have to change that.
I've already done my part, I've strongly advocated the buy of Athlon-based computers for the software developers where I work, and hopefully they'll come (I'm not in IS, but they usually listen to what we developers have to say ;)
So, all you tech-people with a chance to make a statement out there - you _are_ helping AMD in breaking Intels monopoly, right?
(and if you ask me, yes, Athlon is the better CPU also ... if it wasn't, I wouldn't recommend it ...)
it's in my head
There is a very nice review of the new 750Mhz Athlon over at http://anandtech.com
The review states that the new Athlon's external L2 cache does not run at 1/2 CPU speed but rather at 1/2.5 of CPU speed. So L2 cache dependant programs don't perform quite as well on the 750Mhz Athlon as anticipated. According to the review, the processor's FPU marks are (of course) faster than it's 700Mhz brother, but does the extra cost justify the minimal performance increase?
Many people would say no (including me), but remember that the consumers don't care about the internals of the processor, but rather care about the Mhz rating... That's why both Intel and AMD have pushed out 700+ Mhz processors already, even though their previous roadmaps show that they weren't going to push out those processors until Q1 or Q2 of 2000.
Anywho... just to inform you guys about the difference in the L2 cache...
A 750 Mhz Athlon cannot be much better than a 733 Mhz Athlon...? Surely the huge price premium is more to make a profit for those k-rad hax0rs who can go and measure their processor speed in gigahertz...
Can't you see what's happening, though? We're actually seeing AMD for the first time as a competetor rather than an alternative.
Back with the K6, AMD leapfrogged Intel for a moment, but when Intel released the Pentium II a month later, AMD cowered back to it's corner, releasing processors where one could only brag about the price rather than the performance.
But now, AMD's no longer accepting second banana. They release Athlon, leapfrogging Intel. Intel releases Coppermine, catching up to Intel and leapfrogging them in MHz speeds. What does AMD do? They leapfrog Intel again in MHz speeds.
AMD no longer wants to say that they have a "cheap alternative" to Intel. They're finally saying "Screw Intel, we got the faster chip!" And they even have TV Adds now kickin Intel in the butt over it! And for the first time, Intel's actually worried over it!
The important point here is that AMD's fighting back Intel, and they're doing quite a nice job, too! Just because 750 MHz isn't much faster than 733MHz, it does mean that AMD's fighting back!
Why you should always buy just below the state of the art. The 750MHz Athlon isnt that much better than the 600 for most things, certainly not in price/performance terms. That dosent mean I dont want one, just that I cannot justify the expense at this time.
I remember reading a guide for how to build a low cost linux system years ago. The advice was to decide how much power you needed, then how much you are willing to pay. Stage 2 is to wait.
I would suggest Feb might be a good time as the 1GHz part will have been out for a month or so & prices will probably have started to slide.
Most of the K6 motherboards I've seen support at most 5x100MHz for a 500MHz clock speed.
What is 533 anyway? 7x75?
This is just more proof that AMD has finally got something working well for once. Rumours have it that AMD could release 1GHz Athlons whenever they want, but they want to upset Intels "750MHz" day in January (the 10th) by releasing it then (or an 800/900MHz Athlon.
The Athlon is a much better design than the PIII. Now it is 0.18micron it should run cooler too. I would estimate that a 750MHz Athlon was around the equivalent of an 800MHz+ PIII.
Of course, there is the motherboard availability problems, but they are out there now in ever increasing numbers. The technology is old enough now for most of the early problems to have been ironed out.
I would like to see the spec figures for the new Athlon. Should provide interesting reading. I noticed that the FP figures for the 733MHz PIII were an abnormal jump higher than the normal PIII figures. Of course, if you really want power, then get an Alpha!
Of course, AMD has yet to concentrate on the SMP versions of the Athlon. When these arrive, Intel might start looking wistfully at its high end server market.
I am glad for this competition in the 'popular' CPU market, as it really drives prices down. Intel are on the losing end at the moment, but I imagine that they will have their day again. I don't think it will be with Itanium or IA-64 though. AMD have their own 64-bit expensions to IE-32 which they are creating, and the compatability _with speed_ issue might sway a lot of people towards AMDs offering rather than Intels EPIC monstrosity. Shame that such a old unwieldy ISA is still alive though.
I remember about a year and a half ago when Motorola confidently predicted that there would be 1GHz PPCs by the end of 1999. Strange that the one company you wouldn't have thought would have got there are within a hairs breadth of this goal.
I havent bought a computer since 1992, yep all 486s at home. How are they cooling these fast chips it must be more than multiple fans and heat sinks.
Microsoft aggravates my tourettes syndrome.
It's nice to see that AMD is keeping ahead of Intel in the speed wars, but not so nice to see they seem to be pushing up into the same sort of price bracket to do so. One of the GOOD things about the AMD Vs Intel wars was that it was acting to push prices for chips down; I don't really *want* to be forced into buying the latest and greatest chip at enormous expense, just so I can play the latest games; I would much rather have a chip at a price I can afford, that is just below the leading edge
That said, the tiny little trailer that points to the new 553MHz K6 at a reasonable price is a much more welcome sight....
--
-=DaveHowe=-
I havent bought a computer since 1992, yep all 486s at home. How are they cooling these fast chips? it must be more than multiple fans and heat sinks. What do you people have for setups at home?
Microsoft aggravates my tourettes syndrome.
But this year there have been at least three companies that went practically overnight from being the standard that all others were measured against to being rather pathetically looking has-beens. Microsoft is in deep shit. Most people consider 3Dfx thoroughly beaten by nVidia in the current generation of cards, and from what I have seen in their next generation cards nVidia will continue to widen the gap. They are finally beginning to handle enough polygons to give realistic outdoor scenes with "real" trees. What is 3Dfx boasting about? Putting four identical old technology chips on a new board, which wastes memory and requires a separate power line in to the card. A buffer that can give you blurred motion lines? Whopee, that must be fun all of five minutes. And now AMD finally getting everything right, including the timing of the launch.
But even if the upstarts are current media darlings they are still fighting an uphill battle. They have less money to throw around on advertising and continued research, and they must make a lot of profit and contine to win consumer loyalty or they risk falling back into obscurity.
************************************************ ***
Being bitter is drinking poison and hoping someone else will die
Does anyone have any info on any Dual Athlon Boards in the works? I looked around Tom's the other day and only saw the 5 single athlon boards available at the moment.. does anyone know if any manufacturers are developing a dual board? I would LOVE to find one.. 1.5GHz would make me very happy.
//Phizzy
"Most European technology just isn't worth our stealing," -- Former CIA chief James Woolsey, referring to Echelon
In case anyone is interested, here are links to the technical specs pages for the Athlon and the K6-2
--
-=DaveHowe=-
...what with all the problems they've had this year - the fiasco over Camino, then the Processor Serial Number thing (the E.U. is considering banning serialed P3's altogether, of course) and serious shortages of its high end processors.
AMD, on the other hand, has the fastest Mhz processor (good for PR, even if the speed increase is only a few percent) and is selling its lower end processors as fast as it can crank them out.
Could it be that we finally have competition in the processor industry? (faint)
There's a good review of Intel's year at The Register.
Gerv
Gah when will the novelty wear off so we can see these babies used in real life, not in this ridiculous only techies would care circus.
The message on the other side of this sig is false.
Small point folks: 750 mega-hertz is .732 giga-hertz. Please, don't make math errors like this, it looks dumb.
BTW: The light that burns twice as bright actually only burns one-eigth as long. Please stop saying it's half, that's wrong.
I also have a lot to say about other homilies but I'll let it rest.
0.75Ghz == 768Mhz
Nope. Sorry - these aren't megabytes and gigabytes. Mega and Giga mean exactly what they mean - million and billion in this case.
I know that AMD has already released information that they will be making the Professional and Select versions of the Athlon processor, but right now Intel is slipping on the high-end workstation side.
.18 micron processors (which are at least $100 cheaper and their closest Xeon cousin).
News.com had an article stating that Dell and HP will not be using some of Intel's newer Pentium III Xeon processors because the processors cost too much and don't provide that much more performance over the new Pentium III
Currently, AMD has been pushing at the consumer and middle-level workstations, but not enough at the high-end workstation and server market enough. With the Camino stink, the possible EU ban on the Pentium III processors, and the cost/performance problems with Rambus... Intel has opened themselves to losing some of their share in the workstation market.
With VIA producing their new PC133/AGP 4x Athlon chipset, this might help Micron and HP to look more at the Athlon processors as a workstation processor and start pushing those workstations at possibly a lower price than their Pentium III counterparts. If AMD starts working on a more robust and more flexible chipset (read: supports multiple processors, possible support for DDR RAM or Rambus, 64-bit PCI and even on-board Firewire), they might be able to influence Dell to start making Athlon-based Precision workstations or Athlon-based PowerEdge servers (running Windows NT, Linux or even FreeBSD!)
While on the same sort of subject someone pointed out in an earlier comment about how AMD is turning into a competater rather than an alternative. This will be significatly more visible with AMD's 64 bit processor vs IA-64 architecture since AMD for the first time will not just be "copying" Intel's instruction set. Personally I think AMD's plan will go over well for a lot of x86 users wanting to go to 64 bit but not wanting to simply expand the architecture rather than go with a whole new one (though both will be able to run 32 bit x86 code, AMD's chip will do it faster though). The real question is how long will it take AMD to come out with it and how many more delays will Intel put on IA-64 (especially considering the first two IA64 chips, Merced and McKinnley are targeted toward the server market, not the desktop market). So the competition between Intel and AMD will only grow hotter in the years to come (not to mention Rambus vs DDR SDRAM, etc...).
For those who might have missed AMD's presentation at this years microprocessor forum (it talks about stuff like LDT and AMD's 64 bit processor) here is the link: http://www.amd.com/prod ucts/cpg/mpf/pres99/micropforum.html
This is interesting, especially the relative shift in stock prices of AMD vs Intel.
Back when I got my K6/2 (I'm now running a K6/3-400), I resolved that I would support AMD exclusively, until the were within 5% of the market share with Intel. I would then just go with whichever company offered the best product/price match for me at any given time... hopefully, AMD and Intel are always neck and neck with each other, letting neither gain an enormous edge so they have to really scamble to compete and keep up, this is the only way to keep innovation alive...
Speaking of which, does anyone have a favorite source of market-share stats like this ? I'd like to keep up on a day-to-day basis. =)
25% Funny, 25% Insightful, 25% Informative, 25% Troll
Hi, Anyone know of the status of Multiple Athelon MB's? And can the current Athelon's be used on the MP boards? !Google! For seriously lost people
Sorry to deride you; I mean, you are backing AMD in your own way. And that's good. But my view is that the best way to back AMD is the way I did, with my wallet. I have bought not one but two processors for the my home computer from them. My first was the AMD K6-2, 300MHz (and it was *so* sweet, what an upgrade from the DX-4 100MHz Cyrix....), and just recently, a K7-600MHz (with a ASUS K7M motherboard). It is a very nice system.
So, I hope I haven't insulted you, and I am glad you are spreading the good Athlon word. I just think that the best method is to *buy* AMD, and not just spread the word.
"Don't mind me cutting myself on Occam's Razor"
*cough* Flamembait *cough* Meta-Moderation here PLEASE! Pitiful Intel slut...
Hilary Rosen's speech was about her love of money and her desire to roll around naked in a pile of money.
I see a number of posts glorying in the thought of AMD eventually beating out Intel. Yet ask yourselves, you (predominantly) Linux users who claim to value "innovation," who is the current innovator? Is it AMD that is just pushing out processors, which, despite their speed, are not technological leaps forward, or is it Intel that is on the cusp of bringing out Merced, which, combined with RAMBUS technology, sponsored by Intel, represents a virtual processor revolution?
Intel has always produced good products and is on the verge of pushing past the outdated i386 architecture; AMD is, on the other hand, what it always has been, namely, a company which lags 1-2 years behind Intel.
I think the Linux community has a tendency to favor the underdog, regardless of facts or the situation...
Those of you interested in the release of the 533Mhz K6-2...you're welcome. I *just* placed an order for the 500Mhz chip this morning. So, of course, to ensure that the things I buy are immediately outdated, AMD felt it necessary to announce the 533 model today...
Ugh. Obsolete before it even arrives...
I agree with you on this. AMD depends on the techie/OEM market to sell many of their chips. It is basic common sense that this would help them.
By the way, the AMD processors I have owned so far are:
AMD 286/16, AMD 386/25, AMD 486/40, AMD 5x86/133, AMD K6-2/350, and an AMD K6-3/450. I'm holding out on the Athlon until the dual (or quad) processor boards come out. If I'm going to go above 450, I want to be able to put in more than one processor.
--- "So THAT's what an invisible barrier looks like!" - Time Bandits
Unfortunately, when I bought my computer (in August) the Athlon chips were out, but motherboards weren't readily available. Whats the current situation with motherboards? How many options does one have?
i dont display scores, and my threshhold is -1. post accordingly.
i dont display scores, and my threshhold is -1. post accordingly.
Discuss
Can someone point me to a comparison of the K6-2 533 and the Athlon 550? THX
the main benefits of process advancements are:
- Size (K7 with 22M transistors is about the size of a 486 with about 1 or 2M) ;) )
- Frequency, yeah make it smaller and you can make it run faster (K7/750 vs. 486/33
- HEAT, yep, make it smaller and it will need less power, and dissipate less heat
The only problem is that process technology did not scale linearly with the number of transistors CPU makers put on the silicon die. Today's CPUs like K7 or P6 use architectural tricks to attain very high clock speeds (deep pipelining).
So yes, today's CPUs may run 3 or 4x hotter than yesterday's but they perform 20 or 30x better !
FYI, K7 has a maximum heat dissipation of 50 or 60 Watts, that's not enormous when you compare it with the 100 Watts of the Alpha 21264 ...
What I want to know is: WHERE THE HELL ARE THE ATHALON SMP MOTHERBOARDS?!
*grumble*
Playing smart-ass when you don't know your stuff can be very embarassing sometimes.
You should be aware that most of the technologies used in the Merced architecture (which you should be aware is now called "Itanium") were previously pioneered in architectures like PA-RISC. 64-bit architectures in general have been around for a long time, too. RAMBUS was developed outside of Intel by Rambus Incorporated.
Intel is bringing these technologies to the desktop market, which is not a bad thing, but there are no recent indications of any special ability to innovate, as none of these technologies originated with Intel.
The only reason AMD has currently stuck with implementing ia32 clones has been that they have not been in a secure enough situation to do much else (although, they have made some interesting technological advances even in implementing their clones). If they manage to gain a reasonably comfortable position in the market, they will finally be able to take the same kinds of risks.
Based on past performance (cheaper processors, better designs, less bastardly behavior, less bungling), I would much rather have a financially secure AMD in the role of introducing newer technology to the desktop market, rather than Intel. That is why I (and I suspect many other slashdot readers) choose to support AMD.
I think the Linux community has a tendency to favor the underdog, regardless of facts or the situation...
I'm sorry if this sounds harsh, but I think some critics (read: you) of the "Linux community" (read: slashdot posters) have a tendency to blindly take Intel PR material at face value, regardless of the facts of the situation.
DNA just wants to be free...
K7 is faster clock for clock than any x86.
K7 runs at higher clock than any x86.
K7 is cheaper than THE-EVIL-ONE.
x86 is still the most used architecture in the world (and it won't be over soon believe me).
Intel that is on the cusp of bringing out Merced
Merced is 2 years late, and rumours about performance are not good at all. OOCH!
which, combined with RAMBUS technology
DOUBLE OOCH!
sponsored by Intel
god it keeps getting worse !
represents a virtual processor revolution?
Yes I think you got it, it's gonna be more virtual than ever !
I was happy to see the AMD commercial for two reasons:
1. IMO, the TV commercial means that AMD is finally reaching the masses (the people that wouldn't buy anything excpet that "Intel Inside" sticker). Making Intel shake, and continue to lower prices.
2. I finally got AMD's pronunciation of that stupid marketing nickname. Better than pentium, that's for sure.
-d9
So I'm confused. What's faster: a 533mhz K6-2, or a 450mhz K6-3? I have a K6-2 300 that I'm gonna upgrade. AMD is sending me mixed signals...
Why have they brought out a new K6-2 at 533MHz?
Surely they should phase out the K6-2 in favour of the K6 III which has integral level 2 cache, and is therefore a bit faster at the same clock speed.
What am I missing?
I'm out of my tree just now but please feel free to leave a banana.
Anyone know if Abit is going to do an Athlon board? I really like their MBs. I'm going to do a BP-6 for now, but would like to have Athlon as an option down the road.
Interested in XFMail? New XFMail home page.
The Athlon is more than slightly faster. We tested it on some (Maya) renders and it was 25-32 percent faster than a Pentium III at the same clock speed.
Note that this is not some contrived benchmark, this is a real world project with a real world program (Maya renderer). Combined with the relatively low cost and the vaporware-ness of coppermine, the Athlon is very attractive right now.
Unless you need SMP of course. I hope they accelerate the creation of SMP Athlon motherboards. If I was AMD I would be doing everything in my power to assist the motherboard makers in the design and manufacture of them.
Have you forgotten that AMD beat Intel to market with enhanced 3D instructions by several months? The K6-2 was out WAY before the Pentium-III, which was simply a Pentium-II with Intel's version of AMD's 3D-Now! instructions. Happily, AMD didn't try and call this a whole new generation of processor like Intel did.
Why the hell would anyone want Rambus??? It's a crappy technology that perfoms worse than 100MHz SDRAM we have now and costs 5 times as much. Oh wait I know -- it's the Intel's marketing hype. I hope they'll still have a chipset that properly supports SDRAM (not Intel's SDRAM-to-RDRAM converter).
___
If you think big enough, you'll never have to do it.
Who moderated this to "Funny"? Pretty sad, IMHO
I'm about to buy a couple of systems for some novice non-nerds. I was about to get them the boring old Pentium III, but I'd much prefer to buy an Athlon. The catch is, the most important thing is service, since these are newbies.
Gateway doesn't do Athlon. What's the best alternative? Any suggestions?
Can your IM do this?
I swear, every new chip Intel ever released was, "targeted towards the server market." Yet you saw the chip first in high priced desktops. Of course they do this so they can justify the high price of the chip, "Well sure it's expensive, but it's not really for the desktop it's for servers!" Yeah right, when was the last time Intel made a non-mainstream CPU?
I agree totally with you on the price war being a awesome thing for us little people. =] But, I can't say AMD is trying really to screw consumers with the pricing of their K7. I think they are excellent prices. You might forget that AMD is much much smaller and has much much more riding on the K7 then Intel has on the P3. Intel can afford to screw up a bit here and there (820, rambus, etc, etc) They have the market(mind)-share and the spare cash to do this with.
On the other hand, AMD really doesn't. They are taking (took?) a major gamble on their K7 chip. I wish AMD the best of luck with their business, and hope everything pays off for them. I wouldn't mind paying 200$ for a K7-500 or so, because I know the chip is worth it. I've read the articles, reviews, etc. I've also seen it in action, so I know my 200$ would be going to something worthwhile.
On the other hand (I'm on my third hand now) Intel charges more Mhz for Mhz on their chips then the superior (IMHO) K7. Why? Because they can. Because they know people will pay for it. AMD can't afford to think like that.
Anyway...That's just my 2 pennies worth.
I remember about a year and a half ago when Motorola confidently predicted that there would be 1GHz PPCs by the end of 1999
Yeah, I was a little disappointed with Moto's batch of chips, especially that nast "errata" which resulted in the G4 being bumped down by 50MHz to fulfill demand.
Since the introduction of the 601, the PPC has been met or beaten each developmental timeline. This past year, it kinda stalled.
Don't forget Joe PeeCee doesn't know the difference between a G4 and a P3; the P3 has higher clockspeed so therefore "it must be better."
However, Moto is discussing a new 780MHz "G4+" according to The Register. It was originally annouced last month. I can't wait to get my mitts on one of those!
Pope
It doesn't mean much now, it's built for the future.
Lot's of motherboards out now. I got mine here, and they did a great job with everything. But anywhere you look, on the web, should have them.
Generally, the MicroStar, the Fester (from AMD), and the ASUS are considered pretty decent. Main problem with the ASUS is that it isn't supported very well (the American webpage has NOTHING on it about the K7M, only the Taiwan page has info on it). I have heard that the FIC board (SD-11) was just a prototype board that was pushed into production because AMD needed the boards.
As for finding the boards, I haven't had a problem.
"Don't mind me cutting myself on Occam's Razor"
> I remember reading a guide for how to build a low cost linux system years ago. The advice was to decide how much power you needed, then how much you are willing to pay. Stage 2 is to wait.
I agree. I just built a dual Celeron 550 system for $260 ! (I'm a game developer and the extra cpu will be VERY welcome for compiling.)
2 x $35 Celeron 366
2 x $30 Alpha-7 Heatsink/Fan
1 x $130 ABIT BP-6
3D card are another great way to prevent your system from coming obsolete in 6 months. My Voodoo2 really extended the life of my old PPro200. Unreal, and Half-life played just fine. It wasn't until I started playing Thief last month, that the system was starting to chug in some of the larger levels.
Cheers
750/1024 = 0.732 Ghz not .75 :)
Gotta remember that 2^10 stuff
Commencing "bowing down in respect" manouevre on my mark...Mark!
"Don't mind me cutting myself on Occam's Razor"
on the nVidia/3Dfx thing.
The chips used in Voodoo4/Voodoo5 are not old chips. They are new fully 32bit colordepth chips. Voodoo3 was the last of the Voodoos based on the old SST1 technology (which does indeed date back to the Voodoo1) AFAIK.
As for the "blurry lines", antialiasing is a major component of image quality. Making an argument on the correct tradeoff for fillrate and polyrate is a subtle thing. GeForce may have the next generation of Voodoos beat for poly rate (I don't have my GeForce, and you can't believe the numbers any of these companies put out; you have to do your own real-world benchmarks), but Voodoo4/5 are clearly ahead in fill rate.
That, to my mind, puts things up in the air as to which will be the better card in the end. Remember, this depends heavily on the kind of things programmers can figure out to do with the technologies available on the board. Not enough programmers are even using the dual texturing currently available on TNTs and Voodoo2+s
...
I guess I just get tired of the automatic siding with the perceived underdogs. Intel had (but seems to have lost) control of the PC chip market, I fully agree. And things are better pricewise and performancewise than they were three years ago, in terms of relative bang for buck (taking into account the predictable scaling in processor speeds).
But I'm not aware that 3Dfx ever had anything resembling a monopoly in the accelerator market. They made the best high-end boards for a long time, and that dominance may now be up in the air. But I'm not aware they ever screwed customers. The 3D market has always been a favorite example of mine of the power of competition -- a lot of companies seeking to one-up the other on a fairly even playing field. And so far the consumer has been the winner.
He concluded that while the reduced L2 cache speed DOES affect some things, clock rate is way more important. This suprised me, the L2 cache speed reduction affected it very minimally.
Well, this is probably because the Athlons have a rather hefty 128KB of L1 cache...so for most apps your data can easily be held in the L1 (after all, a Celeron has an 128KB L2). Therefore most of the time you're not even going to be needing to access the L2, so the increase in L2 latency won't show up too much.
If the 750MHz Athlon was benchmarked on some server or maybe scientific code (which tends to make more use of the higher-level caches, due to the apps' larger data sets) then we'd probably see a bigger performance hit. The desktop-oriented benchmarks in the Anandtech review, however, don't stress the caches that much.
It's been tested and retested so many times. Even at 66MHz, the memory bandwidth is morethen enough to keep the CPU working at full speed.
___
If you think big enough, you'll never have to do it.
Yeah these CPU's are real impressive but when is anyone going to come out with a real motherboards for these.. Aren't these suppose to support multiprocessing? till then 750 means nothing..
but, you cant get more then 450mhz, now can you?
I'd really like a 10Ghz Athlon, but it just aint gonna happen
--
"Subtle mind control? Why do all these HTML buttons say 'Submit' ?"
ReadThe ReflectionEngine, a cyberpunk style n
Actually, 66MHz (or 16MHz, which is the actual speed of random memory accesses using existing commodity DRAMs, 8MHz if you include the paging table lookup penalty for missing the TLB) is not enough. qv this paper on The Memory Wall and do your own math:
The Memory Wall (postscript)
Remember -- your DRAM isn't operating at 66MHz or 100MHz or 133MHz, except for sequential access. For the purposes of the issues discussed in this paper, sequential accesses are the same as cache accesses. To get numbers relevant to today's PC technology, use the actual random access DRAM latency: 120ns (60ns to look up the paging table, 60 ns to access the addressed memory).
-- Guges --
Binary is set in powers of two (on, off) and hard drives are measured in binary capacity! It's not some obsession, it's how the damn things actually work. Binary is the very basis of this entire technology.
Here's my DeCSS mirror, where's yours?
AMD does have a winner in its Athlon series. Its performance even outweighs setups that utilize dual Pentiums. It even gives the Motorola/IBM G4 a run for the money. According to Barefeats.com, the Athlon holds its own well against a 400 MHz G4 Macintosh (Yosemite motherboard) and a dual-600 MHz PIII WinNT machine.
Bare Feats comparison of Athlon, Dual PIIIs, and G4
-----
Linux user: if (nt == unstable) { switchTo.linux() }
Those who laugh at you for you having a Mac.. are the people who constantly call you to fix their PC.
heh heh, have you actually looked at the sticker price for those babies? OUCH! Sure, they're awesome machines, and if I could afford one, I would grab one in a second. But, sadly, I'm not rich. Neither are most other consumers who buy this stuff. So, we all must stick with x86 for the price. There's more software available for it, anyways.
Here's my DeCSS mirror, where's yours?
infact, its about 10 - 15x as much :)
--
"Subtle mind control? Why do all these HTML buttons say 'Submit' ?"
ReadThe ReflectionEngine, a cyberpunk style n
you don't need a book on multithreading, I used some in a simple sample program I wrote make a function in the format
UINT functionName(LPVOID pParam)
then call AfxBeginThread (functionName,&pParam);
pParam is the paramiter you can pass. If you want you can make it a pointer to a class, and call a 'run' function or similar (its not quite as simple as Java, where you interface runnable, then make a new Thread instance, and start it but its close). Look up the help on AfxBeginThread (Its in afxwin.h)
I don't know how to do it in linux, I think it involves the 'fork()' function, or somthing. but you were talking about MS-flight sim, so I guess you're a windows user
--
"Subtle mind control? Why do all these HTML buttons say 'Submit' ?"
ReadThe ReflectionEngine, a cyberpunk style n
those SPEC95 numbers you wanted to see:
CPU_____________Clock___int95___fp95
Alpha 21264A____700MHz__39.1____68.1
Athlon__________700MHz__31.8____23.8
Athlon__________750MHz__32.8____24.3
Pentium III_____733MHz__31.5____18.0
As an aside, why doesn't Slashdot allow TABLE tags in comments?! It would make the above chart much easier to write/read.
- A.P.
--
"One World, one Web, one Program" - Microsoft promotional ad
"Remember when the U.S. had a drug problem, and then we declared a War On Drugs, and now you can't buy drugs anymore?"
Mhz, Mhz, Mhz... that's good for market only. But what's so exciting about the 0.75GHz fact? Are you "better" if you own "faster" CPU? I think the point is what is innovative -- and it is...hmm....nothing. Intel promises Merced/Itanium architecture -- sounds good in comparison to x86 while AMD and its SledgeHammer pleases Micro$oft with continuing x86 in 64bits....
Then there's the KX133. VIA advertised an early November release date both before and after the earthquake. However, all the Comdex articles point to a January 2000 release of KX133 motherboards.
Not to mention the fact that nobody is advertising SMP KX133 boards, not even the Mighty Tyan, who showcased the Dolphins 2 board at Comdex which featured only one measlely CPU slot.... and this from a company renown for their dual processor high end boards. It's almost as if people *want* the upcomming Athlon Ultra to fail... I mean come on, who in thier right mind would buy a crappy old Xeon with 2 megs of cache for $2k + when you can have an Ultra with much better FPU architecture and *8* megs of cache at a lower cost?
It looks like people are out to screw AMD, but I will support them, and I'm happy that they didn't lose money this quarter, and I hope they clean up next quarter, and sell a ton of Athlons in Europe & the U.S.
I for one would buy AMD even if it were inferior *which it isn't* because I do_not_want a CPU with some nefarious embeded ID # broadcasting across the internet for whatever reason, and you shouldn't either.